Systemic vasculopathy with altered vasoreactivity in a transgenic mouse model of scleroderma

Abstract

Introduction: Vasculopathy, including altered vasoreactivity and abnormal large vessel biomechanics, is a hallmark of systemic sclerosis (SSc). However, the pathogenic link with other aspects of the disease is less clear. To assess the potential role of transforming growth factor beta (TGF-beta) overactivity in driving these cardiovascular abnormalities, we studied a novel transgenic mouse model characterized by ligand-dependent activation of TGF-beta signaling in fibroblasts.

Methods: The transgenic mouse strain Tbeta RIIDeltak-fib is characterized by balanced ligand-dependent upregulation of TGF-beta signaling. Aortic and cardiac tissues were examined with histologic, biochemical, and isolated organ bath studies. Vascular and perivascular architecture was examined by hematoxylin and eosin (H&E) and special stains including immunostaining for TGF-beta1 and phospho-Smad2/3 (pSmad2/3). Confirmatory aortic smooth muscle cell proliferation, phenotype, and functional assays, including signaling responses to exogenous TGF-beta and endothelin-1, were performed. Aortic ring contractile responses to direct and receptor-mediated stimulation were assessed.

Results: Aortic ring contractility and relaxation were diminished compared with wild-type controls, and this was associated with aortic adventitial fibrosis confirmed histologically and with Sircol assay. TGF-beta1 and pSmad 2/3 expression was increased in the adventitia and smooth muscle layer of the aorta. Aortic smooth muscle cells from transgenic animals showed significant upregulation of TGF-beta- responsive genes important for cytoskeletal function, such as transgelin and smoothelin, which were then resistant to further stimulation with exogenous TGF-beta1. These cells promoted significantly more contraction of free floating type I collagen lattices when compared with the wild-type, but were again resistant to exogenous TGF-beta1 stimulation. Aortic ring responses to receptor-mediated contraction were reduced in the transgenic animals. Specifically, bosentan reduced endothelin-mediated contraction in wild-type animals, but had no effect in transgenic animals, and endothelin axis gene expression was altered in transgenic animals. Transgenic mice developed cardiac fibrosis.

Conclusions: The histologic, biochemical, and functional phenotype of this transgenic mouse model of scleroderma offers insight into the altered biomechanical properties previously reported for large elastic arteries in human SSc and suggests a role for perturbed TGF-beta and endothelin activity in this process.

Figure 1

Vascular fibrosis in transgenic mice is associated with increased TGF-β expression and signaling. H&E, Masson trichrome, and picrosirius red (viewed under polarized light) staining of wild-type and transgenic thoracic aorta sections (a-c). Smooth muscle layer architecture, elastic fibers, and smooth muscle layer are normal in the transgenic animals. However, adventitial collagen content is increased, and fibers are thicker in the transgenic animals on Masson trichrome and sirius red staining. LAP(TGF-β1) and free TGF-β1 expression is increased particularly in the adventitia (arrows) but also in smooth muscle cells (d-f). Immunostaining for pSmad2/3 confirms increased nuclear translocation in the transgenic compared with the wild-type. Original magnification, ×20; representative images for panels (a-f) from transgenic (n = 6) and wild-type (n = 6) littermate sex-matched controls. (g) Serial measurements of adventitial and smooth muscle thickness show transgenic adventitial thickening and smooth muscle layer attenuation; summary data are expressed as mean ± SEM. *P < 0.05, from transgenic (n = 6) and wild-type (n = 6) littermate sex-matched controls. (h) Summary data from measurement of non-crosslinked collagen concentration compared with collagen standards (Sircol assay) show significantly higher collagen in transgenic animals compared with wild-type. Data are expressed as mean ± SEM.*P < 0.05, from transgenic (n = 8) and wild-type (n = 8) littermate sex-matched controls.

Figure 2

Altered aortic ring vasoreactivity in transgenic mice. (a) Representative vessel wall tension data from one wild-type (black trace) and two transgenic (red traces) aortic rings, showing attenuated contraction to two concentrations of KCl (30 and 80 mmol/L). (b) Summary data for wild-type (n = 7) and transgenic (n = 9) mice. Transgenic mouse aortae show significantly reduced contractile ability in response to direct stimulation of the smooth muscle cells with KCl at both low and high concentrations. (c) Phenylephrine (PE)-induced contraction is attenuated in transgenic aortic rings, and (d) U46619-induced (a stable thromboxane analogue) contraction is also reduced in the transgenic animal, plotted in response to cumulative concentrations of ligand. Vessels were obtained from WT (n = 8) and TG (n = 5) mice. (e) Vascular responses of aortic arteries isolated from WT (n = 9) and TG (n = 5) mice to cumulative concentrations of sodium nitroprusside (SNP) after preconstriction with U46619, data throughout figure are expressed as mean ± SEM; *P < 0.05.

Figure 3

Cultured vascular smooth muscle cells from transgenic mice have a TGF-β- activated phenotype. (a) Reporter gene assay for β-galactosidase shows equal chemiluminescence in vSMCs from wild-type and transgenic mice. Transgenic fibroblasts from the same animals were used as a positive control. Data are expressed as mean ± SEM; *P < 0.05 from WT (n = 3) and TG (n = 3) animals. (b) β-Galactosidase immunostaining results comparing vSMCs and fibroblasts from wild-type and transgenic animals show negative staining in all vSMCs and positive staining only in transgenic fibroblasts. Images shown are representative of three independent experiments. (c) vSMCs from transgenic mice show increased expression of smoothelin gene by qPCR. Recombinant TGF-β1 induced smoothelin mRNA in wild-type vSMCs, but the response in cells from transgenic mice was attenuated. Immunostaining confirms constitutive overexpression of smoothelin in vSMCs from transgenic mice. (d) Data for transgelin, a second gene important for vSMC cytoskeletal function, show the same trends. Data are expressed as mean ± SEM; *P < 0.05, **P < 0.001; and are representative of three independent experiments examining four littermates for each condition. Original magnification, ×40.

Figure 4

Vascular smooth muscle cells from Tβ RIIΔk-fib mice show enhanced remodeling of floating type I collagen gel lattices. vSMCs from transgenic animals promoted more contraction of free-floating collagen lattices, resulting in gels of reduced diameter and weight when compared with wild-type. Induction by exogenous TGF-β1 resulted in further contraction by wild-type cells, but cells from transgenic mice were refractory to further induction. Data are expressed as mean ± SEM;*P < 0.05, **P < 0.001, and are representative of three independent experiments examining three littermates for each condition.

Figure 5

Perturbed endothelin receptor expression and function in vascular smooth muscle cells from transgenic animals. (a, b) vSMCs from transgenic mice have reduced expression of ETRA mRNA and protein when compared with wild-type cells. Exogenous administration of TGF-β or ET-1 to cells from both wild-type and transgenic animals further suppressed ETRA expression. Data are representative of three independent experiments examining four littermates for each condition and are expressed as mean ± SEM,*P < 0.05, **P < 0.001. (c) Vasoconstrictor response of aortic rings to ET-1 was attenuated in transgenic mice. Bosentan attenuates the response in both wild-type and transgenic mice. Representative data from dose response curves of WT (n = 6) and TG (n = 6) aortic rings before (-) and after (+) response to 10^-8 mol/L ET-1 concentrations, before (-) and after (+) bosentan 2 μmol/L pretreatment. Data are expressed as mean ± SEM; *P < 0.05.

Figure 6

Myocardial fibrosis in Tβ RIIΔk-fib transgenic mice. (a) Transgenic animals have significantly higher non-crosslinked collagen content when compared with sex-matched littermate controls (n = 10). Data are expressed as mean ± SEM, *P < 0.05. (b) Picrosirius red stains of the left ventricles of WT and TG animals viewed under bright-field and polarized light microscopy. Transgenic animals have a diffuse increase in collagen deposition within the myocardial interstitium. Original magnification, ×20, representative images from transgenic (n = 6) and wild-type (n = 6) littermate sex-matched controls.